The United States, Britain and more than a dozen other countries on Sunday unveiled what a senior U.S. official described as the first detailed international agreement on how to keep artificial intelligence safe from rogue actors, pushing for companies to create AI systems that are “secure by design.”
